The cascode circuit is a two-stage circuit where a common base transistor is connected in series with a common emitter transistor as shown in Figure 1. In this simplest application the base of Q2 is held at zero volts (although it can be any fixed voltage as we will see). Q1 is connected as a standard common-emitter amplifier. The cascode topology applies equally to JFETs as it does to BJTs, but we will stick with the latter for the purposes of this discussion.
The cascode topology consists of a common-base amplifier Q2, in series with a common-emitter amplifier Q1. This circuit addresses both the Miller and Early effects in Q1.
Why on earth would we want to do this? Well, cascode circuits help eliminate two of the biggest downsides of BJTs; the Early effect and the Miller effect. The Early effect is the phenomenon where the collector current in a BJT changes as the collector voltage is varied, even if the base is held at a constant voltage. You can see this in Figure 2. Each black curve represents the relationship between VCE andIC for a fixed value of VBE (equivalent to a fixed IB at constant temperature).
The Early effect is an increase in collector current with increasing collector-emitter voltage, with a constant base-emitter voltage. It is the reason for the positive slope of the IC-VCE characteristic in the active region.
The slope of the characteristic curves in the active region is not horizontal as we might expect if IC was depended only on VBE. Instead, there is a positive gradient due to the Early effect. If we extend the gradients to the left as shown in red in the diagram, they will converge at some negative voltage VA, known as the Early voltage.
For a typical transistor with an Early voltage of around 100V we will see a 10% change in IC for a 10V change in VCE. Figure 3 shows a practical example of the Early effect in a current mirror application. The nominal 10mA output of the mirror shown on the left of the figure will therefore change by 40% (from 10mA to 14mA) as the source voltage is varied over the range 10V to 50V, since Q2’s VCE changes with the supply voltage.
The Early effect means the output current in the simple current mirror on the left will change from 10mA to 14mA as VCC ranges from 10 to 50V. The Wilson mirror on the right includes the cascode transistor Q3 which ensures the collector voltages of Q1 and Q2 are fixed, ensuring the output current remains at 10mA over the supply voltage range.
We can fix this by adding a cascode transistor to create a Wilson current mirror as shown on the right-hand side of the figure. Transistors Q1 and Q2 form a current mirror as before, but the collector voltages of Q1 and Q2 are now fixed (at 2VBE and VBE respectively). Q3 supports the supply voltage but is not responsible for determining the current, so its collector voltage change does not matter. The output current will now remain very close to 10mA over the 10 to 50V input voltage range.
The Miller effect is another bugbear of transistor circuits that the cascode can help eliminate. Figure 4 shows a standard common-emitter amplifier with the transistor’s parasitic capacitances shown. It’s easy to see that CBE appears as a capacitance to ground on the input of the amplifier and CCE appears as a capacitance to ground at the output.
The Miller effect means that the collector-base capacitance appears K+1 times larger than its real value (where K is the gain). This can have a huge effect on the circuity bandwidth, particularly if the gain of the stage is high.
The Miller effect (detailed in an earlier article) is another phenomenon that can cause issues with BJT circuits. Figure 4 shows a simple common-emitter amplifier (with bias components eliminated for clarity) along with its stray capacitances. These are usually just a few picofarads each. CBE and CCE appear across the amplifier’s input and output respectively.
The Miller effect means that a capacitance with a value of (1 + K)CCB appears in parallel with CBE,where K is the gain of the amplifier stage. The reason for this magnification is that the small changing voltage on the base end CCB drives a large and opposite changing voltage on the collector end. The current through the capacitor is thus scaled up by a factor of the gain, and so its impedance is scaled down.
This large input capacitance can have a big impact on the bandwidth particularly if the stage has high gain. Adding a cascode transistor as shown in Figure 5 eliminates the Miller effect since the collector of Q1 is now held at a fixed voltage by the cascode transistor Q2. We will still see the capacitances CCB, but the magnification of its gain is eliminated. In this case Q2’s base is held at some fixed voltage by a bias network that is not shown in the diagram.
The cascode topology kills the Miller effect by holding the collector of Q1 at a fixed voltage.
So, we have seen that the cascode topology is a great one for holding the collector voltage of a transistor constant, eliminating the Miller and Early effects – two of the nastier drawbacks of BJT amplifiers.
References
“Cascode.” In Wikipedia, December 21, 2024.
https://en.wikipedia.org/w/index.php?title=Cascode&oldid=1264347124.
Horowitz, Paul, and Winfield Hill. The Art of Electronics. Third edition, 11th printing, with Corrections. Cambridge New York, NY: Cambridge University Press, 2017.
Sponsor this Article
Andrew Levido (andrew.levido@gmail.com) earned a bachelor’s degree in Electrical Engineering in Sydney, Australia, in 1986. He worked for several years in R&D for power electronics and telecommunication companies before moving into management roles. Andrew has maintained a hands-on interest in electronics, particularly embedded systems, power electronics, and control theory in his free time. Over the years he has written a number of articles for various electronics publications and occasionally provides consulting services as time allows.
